CORRELATION OF INCREASED EXPRESSION OF INTERCELLULAR ADHESION MOLECULE-1, BUT NOT HIGH LEVELS OF TUMOR NECROSIS FACTOR-a, WITH LETHALITY OF PLASMODIUM YOELII 17XL, A RODENT MODEL OF CEREBRAL MALARIA

Previous studies demonstrated that Plasmodium yoelii 17XL, a lethal strain of rodent malaria, causes a syndrome in SW mice that resembles human cerebral malaria. The mouse brain pathology is characterized by cytoadherence of parasitized erythrocytes. Here, the possible mechanisms mediating cerebral malaria in this model were studied and the results were compared with a nonlethal strain of this parasite, P. yoelii 17XNL (nonlethal), which does not cause cerebral malaria. Immunostaining for intercellular adhesion molecule-1 (ICAM-1) revealed an increase in expression of this protein in the small venules and capillaries of the brains of infected mice that increased with time after infection. Staining was more pronounced during the lethal infection than the nonlethal infection. Some staining with monoclonal antibody to vascular cell adhesion molecule-1 was also observed, but it was quantitatively less than ICAM-1 staining and was limited to larger venules. During the lethal infection, levels of tumor necrosis factor-a (TNF-a) increased rapidly, peaking on day 4. In contrast, mice infected with nonlethal P. yoelii had a slower serum TNF-a response that peaked on day 10, prior to the maximum parasitemia. In addition, mice with a targeted disruption of the TNF-a gene (TNF-a2/2 mice) were infected with the lethal and nonlethal strains of P. yoelii 17X. The TNF-a2/2 mice infected with the nonlethal parasite had significantly higher levels of parasitemia than controls, whereas TNF-a2/2 mice infected with the lethal strain had slightly higher levels of infected erythrocytes but were equally susceptible to death from this infection. Thus, TNF-a does not appear to be essential in mediating death. These results demonstrate that P. yoelii 17XL infection has features in common with human cerebral malaria and suggest that this model may be useful in testing strategies to alleviate this syndrome. Cerebral malaria refers to the severe complications of falciparum malaria in which vascular plugging of Plasmodium falciparum–infected red blood cells (RBCs) in the brain can lead to coma and death, often in children.1,2 Cytoadherence of parasitized RBCs to endothelial cells, via ligands present on the erythrocyte membrane knobs, is thought to be responsible for this phenomenon.3 However, exceptions to the requirement for knobs have been observed.4 Others have postulated an important role for cytokines and nitric oxide in the pathogenesis of cerebral malaria.5 The molecular mechanisms underlying the cytoadherence of P. falciparum-infected RBCs to the endothelium is under intensive investigation. Most laboratory-maintained and field isolates of infected RBCs appear to bind to CD36. However, binding to intercellular adhesion molecule-1 (ICAM-1) has also been observed.4 Importantly, variation in specificity for endothelial ligands is common among different isolates.6 Vascular cell adhesion molecule-1 (VCAM-1), a member of the same immunoglobulin-like superfamily as ICAM-1, and E-selectin have also been found to be receptors for P. falciparum-infected RBCs in vitro.7 This binding is dependent on activation of human umbilical vein endothelial cells with tumor necrosis factor-a (TNF-a). Thus, the relative importance of the various endothelial cytoadherence receptors for P. falciparum-infected RBCs in the etiology of cerebral malaria may depend on both isolate specificity and cytokine